GLP-1 drugs: fewer falls in older adults

A propensity-matched cohort study published on 11 July 2026 in Osteoporosis International found that semaglutide and tirzepatide were each associated with significantly lower rates of femoral fractures and falls compared with DPP-4 inhibitors in adults aged 65 and over with type 2 diabetes and a BMI of 25 or higher^[1]. The finding adds a new dimension to the GLP-1 class profile at a time when falls and fractures are a leading cause of hospitalisation and functional decline in older adults.

What the study found

Chen and colleagues drew on the TriNetX real-world database for 2018 to 2025, enrolling 27,896 propensity-matched patients in the semaglutide versus DPP-4 arm and 12,808 in the tirzepatide versus DPP-4 arm^[1]. For femoral fractures, the semaglutide group had a 0.3% event rate against 0.5% in the DPP-4 group (hazard ratio 0.488; 95% confidence interval 0.367 to 0.649; P less than 0.0001). The tirzepatide group had a 0.2% event rate against 0.4% (hazard ratio 0.452; 95% CI 0.280 to 0.729; P equals 0.0008).

The fall-risk numbers followed a similar pattern. The semaglutide group had a 3.6% fall rate against 5.4% in the DPP-4 group (hazard ratio 0.663; 95% CI 0.612 to 0.718; P less than 0.0001). Tirzepatide users also had a 3.6% fall rate, against 5.7% on DPP-4 inhibitors (hazard ratio 0.664; 95% CI 0.591 to 0.747; P less than 0.0001)^[1]. The authors noted the benefits were particularly pronounced in participants with a BMI of 30 or above.

Why GLP-1 drugs might reduce falls

The mechanisms are not established with certainty. Several pathways have been proposed in the literature. Weight loss itself reduces mechanical load on joints, which may improve balance and gait stability in older adults. Better glycaemic control reduces episodes of hypoglycaemia, and hypoglycaemia is one of the recognised triggers of falls in older adults with diabetes. GLP-1 receptors are also expressed in bone and muscle tissue, and some preclinical research suggests direct effects on bone metabolism, though the human data here remains limited.

A 2026 review in Drugs and Aging specifically categorises GLP-1 receptor agonists as agents with potentially beneficial skeletal effects and recommends their preference over thiazolidinediones in older patients with type 2 diabetes when choosing antidiabetic agents^[2]. The Chen et al. study provides real-world observational evidence consistent with that recommendation, though it cannot establish cause and effect.

What this study cannot tell us

This is an observational retrospective cohort study, not a randomised controlled trial. The propensity-score matching controls for many confounders, but unmeasured differences between patients who were prescribed GLP-1 agonists and those who received DPP-4 inhibitors cannot be fully excluded. Patients who are prescribed newer weight-loss drugs may differ in baseline health behaviour, physical activity, or access to care in ways that also affect fall risk. The study population had a BMI of 25 or above and was limited to adults aged 65 and over with type 2 diabetes, so the findings do not generalise to younger patients or those without diabetes. For context on how these two drug classes compare on glycaemic control more broadly, the SURPASS-2 trial is the reference point: tirzepatide outperformed semaglutide 1 mg on HbA1c and weight reduction in a 40-week randomised head-to-head^[3].

The femoral fracture event rates in both groups were low in absolute terms (below 0.5%). That makes relative risk reductions look large, and clinicians should read hazard ratios in that context. An absolute risk reduction from 0.5% to 0.3% is clinically meaningful at a population level, but it differs from a scenario where baseline risk is high.

Where this sits in the broader picture for older adults

Falls and fractures are a primary safety concern when prescribing any weight-loss treatment to older adults. Rapid weight loss can accelerate muscle loss, which itself increases fall risk. Earlier research has raised questions about whether GLP-1 agonists might increase falls through orthostatic hypotension or early-treatment gastrointestinal side effects. The Chen et al. data, covering a seven-year observation window, suggests that at a population level the net effect on fall and fracture risk is protective, not harmful, for this patient group. The related question of sarcopenia during GLP-1 treatment in older adults, and what resistance exercise can do to mitigate it, is covered in our piece on GLP-1 drugs and older adults: key evidence.

This article is for educational and journalistic purposes only and does not constitute medical advice. Any decision about antidiabetic therapy, fall-prevention strategies, or fracture risk management in older adults should be made in consultation with a qualified healthcare provider. Medicines discussed are prescription-only in most jurisdictions.